3URMHFWHG,PSDFWRI 5HODWLYH6HD/HYHO5LVHRQWKH 1DWLRQDO)ORRG,QVXUDQFH3URJUDP

Transcription

1 3URMHFWHG,PSDFWRI 5HODWLYH6HD/HYHO5LVHRQWKH 1DWLRQDO)ORRG,QVXUDQFH3URJUDP October 1991 FEDERAL EMERGENCY MANAGEMENT AGENCY FEDERAL INSURANCE ADMINISTRATION

2 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration PREFACE This study of the impact of relative sea level rise on the National Flood Insurance Program was authorized by Congress and signed into law on November 3, The requirements of this study as specified by the legislation are as follows: SEC. 5. Sea Level Rise Study The Director of the Federal Emergency Management Agency shall conduct a study to determine the impact of relative sea level rise on the flood insurance rate maps. This study shall also project the economic losses associated with estimated sea level rise and aggregate such data for the United States as a whole and by region 1. The Director shall report the results of this study to the Congress not later than one year after the date of enactment of this Act. Funds for such study shall be made available from amounts appropriated under section 1376(c) of the National Flood Insurance Act of Discussions with Congress subsequent to the passage of the legislation clarified that the study by FEMA would pertain only to the impact of sea level rise on the National Flood Insurance Program. i

3 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program EXECUTIVE SUMMARY This report contains the findings and conclusions concerning how the National Flood Insurance Program (NFIP) would be impacted by a rise in relative sea level. Based on information recently released by the United Nations on the range in the magnitude of potential rise in sea level, two primary sea level rise scenarios were examined, a 1-foot and 3-foot increase by the year Under both scenarios, the elevation of the 100-year flood would be expected to increase by the amount of the change in sea level. The area inundated by the 100-year flood is estimated to increase from approximately 19,500 square miles to 23,000 square miles for the 1-foot scenario, and to 27,000 square miles for the 3-foot scenario. The region most significantly affected would be the Louisiana coast, where subsidence rates of 3 feet per century would compound the impact of global changes in sea level. Because of potential growth in population within the coastal areas of the Nation over the next century, as well as the expansion of the floodplain, the number of floodprone households is estimated to increase from approximately 2.7 million to 5.7 million and 6.8 million by the year 2100 for the 1- foot and 3-foot scenarios, respectively. Assuming current trends of development practice continue, the increase in the expected annual flood damage by the year 2100 for a representative NFIP insured property subject to sea level rise is estimated to increase by percent for a 1-foot rise, and by percent for a 3-foot rise in sea level. Based on these findings, the aspects of flood insurance rate-making that already account for the possibility of increasing risk, and the tendency of new construction to be built more than one foot above the base flood elevation, the NFIP would not be significantly impacted under a 1-foot rise in sea level by the year For the high projection of a 3-foot rise, the incremental increase of the first foot would not be expected until the year The 60-year timeframe over which this gradual change occurs provides ample opportunity for the NFIP to consider alternative approaches to the loss control and insurance mechanisms of the NFIP and to implement those changes that are both effective and based on sound scientific evidence. Because of the present uncertainties in the projections of potential changes in sea level and the ability of the rating system to respond easily to a 1-foot rise in sea level, there are no immediate program changes needed. However, the possibility exists for significant impacts in the long term; therefore, the Federal Emergency Management Agency (FEMA) should: continue to monitor progress in the scientific community regarding projections of future changes in sea level and consider follow-on studies that provide more detailed information on potential impacts of sea level rise on the NFIP; ii

4 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration in the near term, consider the formulation and implementation of measures that would reduce the impact of relative rise in sea level along the Louisiana coast; and strengthen efforts to monitor development trends and incentives of the Community Rating System that encourage measures which mitigate the impacts of sea level rise. iii

5 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program Acronyms and Abbreviations BFE Base Flood Elevation EPA Environmental Protection Agency FEMA Federal Emergency Management Agency FIA Federal Insurance Administration FIRM Flood Insurance Rate Map FP Floodplain IPCC Inter-Governmental Panel on Climate Change NASA National Aeronautics and Space Administration NFIP National Flood Insurance Program NGVD National Geodetic Vertical Datum of 1929 NOAA National Oceanic and Atmospheric Administration NRC National Research Council PGR Post Glacial Rebound SFHA Special Flood Hazard Area SWFL Stillwater Flood Level iv

6 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration Conversion Table -- English to Metric Units Multiply By Obtain Inches (in) Millimeters (mm) Centimeters (cm) Feet (ft) Centimeters (cm) Meters (m) Miles (mi) 1.61 Kilometers (km) Square Miles (mi 2 ) 2.59 Square Kilometers (km 2 ) Temperature Change [Degrees Fahrenheit ((F)] 5/9 Temperature Change [Degrees Celsius ((C)] To obtain absolute Fahrenheit (F) temperature readings from Celsius (C) readings, use formula: F = 9/5 C + 32 v

7 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program TABLE OF CONTENTS 1.0 Summary 1 Page 1.1 Background Study Objectives and Approach Findings Conclusions and Recommendations Introduction Sea Level Rise in the United States Purpose of Study Physical Changes Methodology Results Demographics Methodology Results Economic Implications for the NFIP Background Methodology Impact on Insurance Premium Requirements Impact on Losses Program Impact Study/Mapping Requirements 57 References 59 vi

8 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration LIST OF TABLES Table 3.1 Value of Coefficient b for the Scenarios 21 Considered in this Report Table 3.2 Milestones for 3-Foot Sea Level Rise 23 Scenario Corresponding to Successive 1-Foot Increments of Rise Table 3.3 Milestones for 1- and 3-Foot Relative 25 Sea Level Rise Scenarios for Louisiana Table 3.4 Area Affected Due to a 1-Foot Rise in Sea Level 34 by the Year 2100 Table 3.5 Area Affected Due to a 3-Foot Rise in Sea Level 35 by the Year 2100 Table 4.1 Estimated Total Households in the Coastal 41 Floodplain for the 0-, 1-, and 3-Foot Sea Level Rise Scenarios by the Year 2100 (In Millions) Table 4.2 Estimated Number of Households by Region in the 42 Coastal Floodplain for the 0-, 1-, and 3-foot Sea Level Rise Scenarios by the Year 2100 (In Thousands) Table 5.1A Post-FIRM Actuarial Increase in Average Premiums 49 for Buildings Subject to Sea Level Rise Required to Maintain Actuarial Soundness Table 5.1B Pre-FIRM Subsidized Increase in Average Premiums 54 for Buildings Subject to Sea Level Rise Required to Maintain Current Subsidy Level Page vii

9 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program LIST OF FIGURES Page Figure 3.1 Eustatic Sea Level Rise for the 1- and 22 3-Foot Sea Level Rise Scenarios Figure 3.2 Major Geomorphic Regions that are Representative 24 of the Range of Subsidence Rates in Coastal Louisiana Figure 3.3 Schematic Diagram of the Effect of Sea 30 Level Rise on the 100-Year Coastal Floodplain Figure 3.4 Coastal Physiographic Regions of the United 32 States Based on Morphological and Geologic Variations Figure 4.1 Population as a Function of Time Based on 39 Information Supplied by Woods & Poole, 1990 Figure 5.1A 1990 Representative Pre-FIRM Distribution 46 (V-Zone) Figure 5.1B 1990 Representative Pre-FIRM Distribution 47 (A-Zone) Figure Representative Post-FIRM Distribution 48 Figure Foot Sea Level Rise Scenario: V-Zone 50 Figure Foot Sea Level Rise Scenario: A-Zone 51 Figure Foot Sea Level Rise Scenario: V-Zone 52 Figure Foot Sea Level Rise Scenario: A-Zone 53 viii

10 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration PROJECTED IMPACT OF RELATIVE SEA LEVEL RISE ON THE NATIONAL FLOOD INSURANCE PROGRAM 1.0 SUMMARY 1.1 Background The rise of global sea level over the past century has been documented by several investigators using tide gage measurements. At specific locations, the change in sea level relative to the land is dependent upon the effects of any local land subsidence or uplift. For areas experiencing a significant rate of uplift (such as portions of the Alaskan coastline), relative sea level has been decreasing, while for areas in which subsidence is taking place (such as portions of Louisiana's coastline), relative sea level is increasing at a more rapid rate than in other areas. Although it is known that mean sea level fluctuates over long time periods, the exact causes of these natural changes are not well understood. It has been suggested by some that the recent rise of sea level is related to global warming (the greenhouse effect) and that, as the atmosphere warms, the oceans will rise because of the melting of ice masses and thermal expansion of the oceans. The magnitude of historical global warming, its anthropogenic and/or natural origins, and its link to sea level rise are issues that are currently subject to intense scientific scrutiny. The potential magnitude of warming and the degree to which it would be delayed by the thermal inertia of the oceans are uncertain. Also, the degree to which changes in precipitation affecting the ice caps and mountain glaciers might change the volume of water removed from the sea and stored is uncertain. The atmosphere and ocean are complex systems, making long-term climate and sea level predictions extremely difficult. Numerical models of global climate change, although ever advancing, are still limited in their ability to accurately predict changes in the atmosphere and ocean over long (decadal or centennial) time scales. Even though these limitations exist, the most sound basis for predicting changes in this global system is the combination of numerical modeling and analyses of the available long-term environmental records (the past 2-8 million years of the geologic record). The above uncertainties have prompted investigators interested in quantifying the impacts of potential sea level rise to address a range of sea level rise scenarios. These scenarios generally assume that the rate of sea level rise will accelerate with time and that a greater rate of rise will occur in the 1

11 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program latter half of the next century. A significant increase in relative sea level could cause extensive shoreline erosion and inundation. Higher relative sea level would elevate flood levels and therefore require alteration of the 100-year coastal floodplain delineated by the Federal Emergency Management Agency (FEMA). Flood events would impact more property and result in greater damage as sea level increased. This problem is exacerbated by the present trend towards increased concentration of population in coastal areas. 1.2 Study Objectives and Approach The primary objective of this study is to quantify the impacts of sea level rise on (1) the location and extent of the U.S. coastal floodplain, (2) the relationship between the elevation of insured properties and the 100-year base flood elevation (BFE), and (3) the economic structure of the National Flood Insurance Program (NFIP). The coastal floodplain area affected includes areas subject to increased erosion and submergence. In response to sea level rise, changes will occur in the extent of the coastal floodplain, in the portion of the coastal floodplain that is subject to flooding and modest wave action (A-Zone), and in the portion of the coastal floodplain subject to flooding and significant wave action (the velocity zone or V-Zone). Areas affected by flooding (both coastal and riverine) are shown on Flood Insurance Rate Maps (FIRMS) published under the NFIP. For this study, an average insurance risk was identified based on flood-depth distributions reflected in current flood insurance policies. Different distributions were assigned to pre-firm and post-firm structure categories. For the purpose of this study, pre-firm structures were defined as structures built before 1980; post-firm structures are structures built after Two sea level rise scenarios for the period 1990 to the year 2100 were examined in this study. Based on recent scientific investigations, the first scenario is a 1-foot rise in sea level by the year The second scenario is the high scenario of a 3-foot rise in sea level by the year Studies supporting these scenarios include the report entitled Scientific Assessment of Climate Change prepared for the Intergovernmental Panel on Climate Change (IPCC) by Working Group No. 1 (IPCC, 1990). The IPCC was jointly established by the World Meteorological Organization and the United Nations Environment Programme. For comparison purposes, a no-rise scenario is also cited in this report. Accomplishing a study of this scope and magnitude required that several assumptions be made. It is important to understand that these assumptions can significantly influence the quantitative results of this study. The major assumptions are described below: 1. Census data were used to establish population trends in each coastal county to the year 2

12 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration Projections beyond 2010 (to the year 2100) were based on these trends. This approach assumes a linear increase of population over time and does not account for development saturation that may occur or other factors which could significantly affect the population trends adopted for this study, such as changes in mortality rates, fertility rates, and social and recreational trends. 2. Within each coastal county, the total households (based on population estimates) were assumed to be uniformly distributed over the total land area of the county. The number of households in the county's floodplain was determined by multiplying the total number of households by the ratio of floodplain area to total county land area. This assumption was necessary because of the lack of quantifiable information about the variation of the density of households in the floodplain. This assumption could lead to either an overestimate or underestimate of the number of floodplain households in each county. 3. This study assumes that no engineering solutions or land use/coastal zone management practices are implemented over the study period other than current practices related to elevation of structures. Options that could substantially mitigate the impacts of sea level rise in open coast areas include armoring of the shoreline (e.g., constructing seawalls, breakwaters, and dikes), beach renourishment, and the adoption of setback regulations. The effect of this assumption is that the projections contained in this report will be overestimated. 4. The obsolescence of structures was not considered in this study. Based on the expected life of a coastal structure, a certain fraction of these structures will become obsolete each year and will be replaced by new structures which will be in compliance with the current NFIP regulations for construction at that time. Since obsolescence has not been accounted for, the actual insurance risk may be overestimated in this study. 1.3 Findings The current total 100-year coastal floodplain area is approximately 19,500 square miles (50,500 square kilometers) for all coastal regions of the United States. Most of this area is contained in the coastal states from the Mid-Atlantic region to the Gulf of Mexico region. The west coast, Alaska, and Hawaii together account for no more than 5 percent of the total coastal floodplain area. The additional areas that may be affected by the 100-year flood are estimated to be approximately 2,200 square miles (5,700 square kilometers) for the 1-foot scenario and 6,500 square miles (16,830 square kilometers) 3

13 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program for the 3-foot scenario when subsidence in Louisiana is not taken into account. When subsidence in Louisiana is accounted for, these figures become 3,400 square miles (8,800 square kilometers) and 7,700 square miles (19,900 square kilometers), respectively. The estimated total number of households in the coastal floodplain for the 1-foot and 3-foot sea level rise scenarios for the year 2100 are shown in the following table. The numbers in brackets reflect the case when subsidence in Louisiana is taken into account. For comparison purposes, expected results for a no-rise condition (i.e., 0-foot scenario) are also shown to indicate the influence of population estimates on the number of floodprone households. Total Estimated Households in the Coastal Floodplain (In Millions) Current Households 0' Scenario 2100 Households in A-Zone [4.6] 1' Scenario [5.1] 3' Scenario [6.1] Households in V-Zone [0.58] 0.61 [0.64] 0.73 [0.75] Total Households in Coastal Floodplain [5.2] [5.7] [6.8] A model representing the shifting distribution of risk characteristics of NFIP business was created to provide some insight into the relative changes in expected losses and resulting premiums caused by an increasing flood risk over time. The analysis was limited to the consideration of the standard flood insurance coverage provided to buildings insurable under the NFIP and not the additional erosion benefits afforded by the Upton-Jones Amendment, which was enacted in The Upton-Jones program and its associated benefits were not considered in this study for several reasons. Engineering solutions, coastal zone management practices, and other options discussed in Item 3 on page 3 would influence the vulnerability of structures and their eligibility for benefits under the Upton-Jones Amendment. Although these kinds of impacts have been investigated in some studies (National Research Council (NRC), 1987; Environmental Protection Agency (EPA), 1989), the effect of sea level rise on the Upton-Jones program cannot be determined without 4

14 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration conducting a study that specifically addresses this issue. Furthermore, even without the additional impacts of sea level rise, there are concerns about the pricing of Upton-Jones coverage and the lack of a companion erosion management program that make the long-term continuance of the present Upton-Jones program problematic. Since this study was undertaken, a bill has been introduced to repeal the Upton-Jones flood policy benefit and substitute a mitigation assistance program under which limited funding would be available for relocation of structures threatened by coastal erosion. In assessing the potential impact of sea level rise, this study examines the sensitivity of the NFIP's rate structure to the changing conditions as an indication of the degree to which program changes would have to be made and of the criticality of the timeframe in which such changes might be needed. A rising sea level in combination with increasing population will not only increase losses, but also increase the number of policies and thus premium income available to pay losses. Therefore, the analysis focused on whether existing rate structures will be adequate to address the problem of maintaining an overall premium income level commensurate with the level of losses, and how premium charges should be distributed among the policyholders who have varying degrees of risk exposure. Because the program will be insuring a dwindling number of pre-firm buildings over the course of 110 years, sea level rise is mainly an issue for post-firm construction. The following table shows the results of the analysis for this latter category of business. Post-FIRM Actuarial Increase in Average Premiums For Buildings Subject to Sea Level Rise Required to Maintain Actuarial Soundness 1-Foot Rise 3-Foot Rise A-ZONE V-ZONE Full Risk Premium Rate Percent Full Risk Premium Rate Percent Change Change % % % % The percent change shown in this table reflects how the average full risk premium rates per $100 of coverage, and therefore the total premium income, for post-firm policies subject to sea level rise would have to increase in order to cover flood insurance losses. The relative change and magnitude of the rates indicate that there is ample flexibility in the NFIP rate structure to accommodate a 1-foot 5

15 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program rise in sea level. A 3-foot rise may require that additional measures be taken to distribute premium burdens equitably and avoid undue cross subsidies. In addition, the potential map revision and restudy requirements were considered. It is estimated that a total of 283 counties will be affected by increases in sea level. For these counties, approximately 5,050 FIRM panels will need to be revised as sea level rises. The cost of revising the affected map panels to account for each 1-foot increase in sea level is estimated to be $30,000,000. This cost would be spread over a 4- to 5-year period. 1.4 Conclusions and Recommendations There is a great deal of uncertainty in the current projections of the rate of sea level rise. Moreover, the aspects of flood insurance rate-making that account for the possibility of increasing risk, and the tendency of post-firm construction to be built more than 1 foot above the BFE combine to eliminate any immediate threat from sea level rise to the NFIP's ability to insure against flood losses through a system of pricing that is fair and that protects the NFIP's financial soundness. There is no need for the NFIP to develop and enact measures now in response to the potential risks that would accompany increasing sea levels. As more information is collected over the next several decades, our ability to analyze past trends and our confidence in predictions will increase, allowing us to better assess both the magnitude of the problem and the most appropriate responses. The high projection of a 3-foot increase by 2100 shows that a 1-foot increase would not be realized until This 60-year horizon provides ample time to consider alternative approaches and implement those that are both effective and based on sound scientific evidence. For these reasons, the following technical and policy procedures are recommended: 1. FEMA must continue to monitor progress made by the scientific community in improving the reliability of projections of the potential increase of relative sea level. A formal report should be prepared beginning in 1995, and every five years thereafter, by the Federal Insurance Administration (FIA) identifying the advances made in the capability to predict potential changes in global sea level. In addition, alternative fiscal and mitigation measures designed to minimize the impact of future increases of sea level on the NFIP should be examined. 2. Because of the more immediate threat and definitive trend of subsidence along regions of the Gulf of Mexico coastline, especially within and near Louisiana, FEMA must explore and consider adoption and implementation of appropriate measures to mitigate the effects of this increasing risk. The process of identifying appropriate mitigation 6

16 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration measures and the data needed to support these measures should include coordination with other Federal and State agencies involved with this problem. The measures implemented in these regions will serve as models for other coastal areas when the broader issue of global change of sea level requires directed action by the NFIP. 3. In the near term, FEMA will increase its efforts to encourage, through the NFIP's Community Rating System, voluntary adoption and enforcement at the State and local level of mitigation measures, such as BFE freeboard requirements and construction setbacks, that take the potential for increases in relative sea level into account. In addition to working at the State and local level, FEMA must also continue its work with national building code organizations to reflect appropriate risks associated with the possibility of rising sea levels. 4. FEMA will continue, and strengthen, its monitoring of the trend of development patterns related to zoning and density to ensure that as trends change there is no degradation that would compromise fundamental goals and objectives of the NFIP. Furthermore, a concerted effort must be made to continue to monitor redevelopment as structures reach the end of their useful life to ensure compliance with minimum NFIP standards. 5. Improvement of this study in the future will depend on the availability of more complete and accurate data and on the ease of manipulating these data. For example, the creation of digital databases of topographic and demographic information would offer the possibility of efficiently computing the physical impacts of sea level rise. These tools would allow for regional or county studies to be performed with more detail and confidence. Also, FEMA could more confidently make projections of potential flood losses. 6. FEMA may undertake in the future a broad-based study to gather and collate shoreline erosion information on a national basis. The results of this effort would permit very site-specific determinations of potential land loss due to sea level rise and would link FEMA's efforts in this area with those of other Federal agencies, e.g., U.S. Geological Survey (USGS) and the EPA. These data would be useful for the judicious implementation of construction setbacks. 7. FEMA should undertake joint studies with other Federal agencies which are involved in the global warming/sea level rise issue, e.g., EPA, USGS, National Oceanic and Atmospheric Administration (NOAA), National Aeronautics and Space Administration (NASA). The capability of FEMA to provide flood loss figures is attractive to the other 7

17 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program agencies that are interested in quantifying the losses or damages associated with sea level rise. 8

18 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration 2.0 INTRODUCTION A rise in sea level could potentially have a major impact on the coastal areas of the United States. Physical effects associated with higher sea level are the inundation of coastal lowlands, increased shoreline erosion, and loss of wetlands. The loss of wetlands will affect the hydrodynamics and therefore the flooding characteristics of tidal bays and rivers. Shoreline recession and submergence of dry land are direct responses to rising sea levels. The most vulnerable areas are coastal wetlands. A 1-meter (3.3-foot) rise in sea level by the year 2100 could result in the loss of percent of the United States coastal wetlands (Titus et al., 1989). The greatest losses are projected to be in Louisiana, where shoreline erosion and land loss rates are presently the highest in the country. Since wetlands act as buffers to the inland penetration of coastal flooding, the loss of these areas will increase the extent and severity of flooding in many areas. An increase in the severity of coastal flooding due to sea level rise and a subsequent increase in shoreline erosion could present a potential hazard for coastal development. Research shows that a significant portion of the Nation's shorelines are currently eroding. Presently, over 70 percent of the world's coastlines are eroding (Bird, 1985). The National Shoreline Study by the United States Army Corps of Engineers (1971) reported that 43 percent of the shorelines in the United States are experiencing erosion. Leatherman (1988) estimated that 90 percent of the U.S. shoreline consisting of sandy beaches is eroding. The average erosion rate, i.e., shoreline retreat, along the Atlantic coast is 2.6 ft/yr (0.8 m/yr) (NRC, 1987). The Pacific coastline has localized areas of erosion. For example, San Diego and Los Angeles Counties have ongoing beach renourishment projects. Erosion in California is episodic and fluctuates according to climatic cycles of storm activity (see, for example, the October 1989 issue of Shore and Beach, Vol. 57, No. 4, which describes in detail the impacts of the January 1988 storm). However, the U.S. Pacific shoreline is considered to be relatively stable since the majority of the coastline is hard rock (NRC, 1987). In the United States, shorelines are retreating because of both natural and man-induced causes. Some scientists suggest that there is a direct causal relationship between landward shoreline retreat and relative sea level rise, which results in the displacement of the shoreline and, in some cases, barrier island submergence (Leatherman, 1983, 1988; Everts, 1985). An increase in sea level will result in higher surge elevations and consequently higher waves. The overall result will be an increase in damage to coastal structures as sea level rises and the severity of storm-induced flooding increases. Coastal structures are increasingly being threatened due to shoreline retreat. Along the Atlantic coast, residential and commercial buildings and erosion control structures are damaged or destroyed each year by moderate northeasters and tropical storms. These structures will be affected to varying 9

19 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program degrees by a rise in relative sea level. As shorelines retreat, larger wave heights are possible due to deeper nearshore waters, resulting in increased wave power and greater destructive force (NRC, 1987). Structures currently designed to withstand a 100year storm event could be overtopped and/or destroyed. Similarly, some buildings that were built above the current BFE would be subject to flooding from such an event. Estimates of the magnitude of sea level rise vary widely within the scientific community. To assess the possible impacts associated with a rise in sea level, the change in sea level must be established. The following sections discuss the findings of various investigators and present both current and historical rates of change. While most experts agree that sea level is rising, opinions differ about the cause and magnitude of the rise. These issues are also briefly discussed in the following sections. 2.1 Sea Level Rise in the United States Scientists recognize and define two types of sea levels: eustatic and relative. Eustatic sea level refers to the global or worldwide height of sea levels. Changes in eustatic sea level result from a number of physical processes, primarily the melting of polar ice masses, thermal expansion of the oceans, and changes in oceanic volumes due to glacial displacement. Relative sea level refers to the height of sea level as measured from the ground at a particular point or area on the earth's surface. Change in relative sea level usually results from the interaction of two different and essentially independent processes: 1) local change (uplift or subsidence) in the absolute elevation of the land mass and 2) change in the absolute elevation of the earth's ocean (eustatic changes). Subsidence is caused by a localized downward displacement of the land mass and can usually be attributed to a number of factors, including 1) tectonic downwarping of the earth's crust, 2) consolidation and compaction of sediments, and 3) withdrawal of subsurface fluids. It is important to note that given fixed eustatic sea levels, subsidence alone could account for dramatic rates of shoreline retreat and increased coastal erosion. For example, in the Teche basin of Louisiana, subsidence rates average 1.11 cm/yr (0.44 in/yr) which accounts for more than 80 percent of the local relative sea level rise for this region (Ramsey and Penland, 1989). Uplift of the land surface is primarily caused by 1) tectonic uplift due to the movements of the earth's ocean and/or continental plates and 2) isostatic rebound, that is, uplift of the continental crust due to the retreat of the glaciers that covered the northern portion of the United States (and Canada) during the end of the Pleistocene epoch, approximately 15,000 years ago. The southeast coast of Alaska is an area which has been experiencing uplift of the land mass. Here, the rate of relative sea 10

20 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration level rise ranges from -2.2 to mm/yr (-0.09 to in/yr) (Gornitz and Kanciruk, 1989), where a negative sign indicates that relative sea level is decreasing. The primary method of measuring rates of sea level rise is to compare historical and recent sea level data that have been collected from tide gages. Unfortunately, accurate long-term tide gage data are unevenly distributed spatially and temporally. For example, the majority of tide gage stations are located in the northern hemisphere, and the longest records are generally for areas in the North Atlantic. Analysis of the tide gage data shows that the rates of relative sea level rise are unevenly distributed across the globe. For example along the southeast coast of Alaska, geologic uplift associated with isostatic rebound is greater than eustatic sea level rise, thus the net result is a localized decrease in relative sea level. Conversely, in the Gulf of Mexico, the extraction of subsurface fluids has caused a decrease in the elevation of the land mass. This decrease, combined with eustatic sea level rise, results in a rapid rise in relative sea level. Tide gage measurements reflect relative changes in sea level; thus to isolate eustatic changes, the effects of uplift and subsidence of the land surface must be removed from the data 2. Most scientists agree that eustatic sea level is rising. Prevailing theories attribute the rise to the combined effects of melting polar ice caps and thermal expansion of the oceans, processes that have been occurring since the glaciers retreated from the northern hemisphere during the end of the most recent Ice Age (Wisconsin Stage of the Pleistocene epoch), about 15,000 years ago. Prior to the decay of the Wisconsin Stage glaciation, sea level was approximately 400 feet lower than present. From 15,000 to about 6,000 years ago, eustatic sea level rose, on average, 3.5 ft/century (1.1 m/century). During the past 6,000-7,000 years, however, the rate of sea level rise has decelerated, and in the past century, global eustatic rise, based on historical tide gage data, has been estimated to range from 1.1 to 3.0 mm/yr (0.04 to 0.12 in/yr) (Carter, 1988). Douglas (1991) suggests that the wide discrepancies among estimates of regional trends of sea level rise are mostly due to the location of tide gage stations on convergent plate boundaries. The resulting contribution of vertical crustal movements due to post glacial rebound (PGR) can account for as much as 50 percent of the observed relative sea level rise (Gornitz et al., 1990). Peltier and Tushingham (1989) examined tide gage records and estimated that global eustatic sea level rise is 2.4 mm/yr (0.09 in/yr) by determining a correction value for PGR at each station. Using the same correction values for PGR established by Peltier and Tushingham (1989), Douglas estimated global eustatic sea level rise to be 1.8 mm/yr (0.07 in/yr), which is comparable to Peltier and Tushingham's 2 Another factor that must be considered and compensated for is the unevenness of the surface of the ocean, caused by the effects of currents, winds, tides, and changes in atmospheric pressure. 11

21 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program estimate (Douglas, 1991). The difference was attributed by Douglas to be due to his exclusion of tide gage records located at convergent plate boundaries. This research demonstrates that a reliable estimate of sea level rise based on tide gage records can not be made without considering PGR (Douglas, 1991). Many scientists predict that the rate of rise will increase in the future due to elevated global temperatures caused by increased levels of greenhouse gases in the atmosphere. The major contributors to greenhouse warming are carbon dioxide (55 percent), chlorofluorocarbons (24 percent), methane (15 percent), and nitrous oxide (6 percent) (IPCC, 1990). The NRC (1983) estimated that there is a 75- percent probability that carbon dioxide concentrations will double by the year With an estimated temperature increase of 1.5( to 5.5( Celsius (C) (2.7( to 9.9( Fahrenheit (F)) associated with an increase in greenhouse gases equivalent to a doubling of CO 2, global mean sea level is expected to rise over the next century. It should be noted, however, that recent studies suggest that global warming due to increased atmospheric concentrations of greenhouse gases may be overstated. For example, in a recent study, Lindzen (1990) analyzed time series for annually averaged surface temperatures dating back to He found that there was no significant variation of global temperatures in excess of 1(C (1.8( F). According to his results, temperatures have been fairly stable during the past 135 years, suggesting that current models may overestimate global warming. A chronological review of recent scientific literature pertaining to global warming scenarios and corresponding sea levels shows the variability in estimates of projected sea level rise. For example: Revelle (1983) estimated that sea level could rise a total of 70 centimeters (2.3 feet) by the year 2085, with a 25 percent margin of error indicated. Hoffman et al. (1986), in an update to Hoffman et al. (1983), predicted future sea level rise in the year 2100 to be within the range of 57 to 368 centimeters (1.9 to 12 feet). Robin (1987) forecast a rise in sea level of 0.8 meter (2.6 feet), with a range 0.2 to 1.6 meters (0.6 to 5.2 feet), by the year A report issued by the NRC (1987) entitled Responding to Change in Sea Level: Engineering Implications included a discussion on mechanisms affecting sea level. The NRC report summarized earlier studies and concluded that a realistic estimate of sea level rise associated with increased carbon dioxide concentrations is from 0.5 to 1.5 meters (1.6 to 4.9 feet). MacCracken et al. (1989) used the oceanic heat transport model of Frei et al. (1988) and estimated less than a 0.5- to 1-meter (1.6- to 3.3- foot) rise in sea level by the year Meier (1990) reports that the current "best estimate" of sea level rise is 0.3 meter (1 foot) by the year 2050, with a "high estimate" of as much as 0.7 meter (2.3 feet) by the 12

22 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration year 2050, and a "low estimate" near zero. The NRC (1990) summarized recent findings on the effect of atmospheric temperature change on the world's oceans. They concluded that "one hundred years from now, it is likely that sea level will be 0.5 to 1 meter (1.6 to 3.3 feet) higher than it is at present." A study prepared by the IPCC, entitled, Scientific Assessment of Climate Change (IPCC, 1990), presented 1-foot, 2.2-foot, and 3.6-foot scenarios as the low, best, and high estimates of sea level rise expected by the year The IPCC was jointly established by the World Meteorological Organization and the United Nations Environment Programme in 1988 to assess scientific information related to various components of the climate change issue. The estimates cited above correspond to a "business-as-usual" scenario; that is, it is assumed that no steps are taken to limit greenhouse gas emissions. Other scenarios were considered in which progressively increasing levels of controls reduce the growth of emissions. These latter scenarios lead to smaller projections of the sea level rise than the "business-as-usual" scenario. Potential contributors to sea level rise include thermal expansion, the Alpine and Greenland glaciers, and the Antarctic Ice Sheet. It is controversial whether the contribution of the Antarctic Ice Sheet to sea level is negative or positive. There is no conclusive evidence to date that shows this ice sheet has contributed to sea level rise over the past 100 years (IPCC, 1990). An increase in global temperatures could increase snowfall accumulation over the sheet, resulting in a negative contribution to sea level. On the other hand, increased temperatures might eventually cause an instability of the ice sheet with outflow of ice and meltwater into the ocean and a rise of sea level. Meier (1990) suggests that much of the meltwater from the polar ice caps will percolate and refreeze in the subfreezing snow. Furthermore, it is unlikely that any contribution from the ice shelves will have an appreciable impact on sea level by the year 2050, given the slow response of the ice shelves to slight changes in global temperature. There is some speculation that this outflow could become significant beyond the 110-year time frame addressed in this report (IPCC, 1990). However, there is great uncertainty on this issue. The IPCC (1990), in formulating its sea level rise scenarios, considered that even in the worst case (high scenario) there would be no contribution from Antarctica. Because of the number of physical parameters involved, it is not possible to assign to the various sea level rise scenarios statistical confidence intervals in a strict sense. The IPCC generated three projections -- best estimate, high, and low -- based on an estimated range of uncertainty in each of the potential contributing factors and in the resulting global warming predictions. If global temperatures increase, changes in climate could occur that would affect hurricane activity. There has been scientific speculation about the effect of global warming on the frequency, 13

23 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program intensity, and tracks of hurricanes. Some scientists theorize that storm frequency and intensity may increase and storm tracks may be displaced farther to the north as global temperatures increase. According to the IPCC (1990), the ocean area having the critical temperature at which tropical storms are created (26( C/79( F) will increase as global temperatures change. However, climate models to date give no indication whether the intensity and frequency of tropical storms will increase or decrease as the climate changes (IPCC, 1990). Mid-latitude storms may consequently weaken or change their tracks in response to warmer temperatures in the northern hemisphere. There is some evidence of a decrease in the irregularity of mid-latitude winter storm tracks based on model simulations (IPCC, 1990). These are research topics that are currently being investigated, and no firm conclusions are available. The effects of a change in climate on precipitation patterns and smaller scale disturbances are continuing to be researched. If these effects are proven to be significant, then there could be an appreciable impact on the characteristics of the 100-year floodplain delineated by FEMA. Because of the uncertainty in the current estimates of future storm patterns due to global warming, no attempt has been made to include these effects in this study. Several studies have examined the local effects of relative sea level rise based on projected increases in sea level. They include the following: Kana et al. (1984) used a concept called "drowned-valley" to project new shorelines based on pre-existing contours for the City of Charleston, South Carolina. Leatherman (1984) projected current shoreline changes along southeast Galveston Bay, Texas, for the years 2025 and Gibbs (1984) performed an economic analysis of the effects of sea level rise on the coastline of the City of Charleston, South Carolina, and the City of Galveston, Texas. In this study, Gibbs examined anticipated losses in dollars due to shoreline retreat and increased inundation. Titus et al. (1991) projected the nationwide economic and environmental impact of sea level rise to the year 2100 in terms of inundation, shoreline retreat, and the costs of protecting developed areas. Because of the high cost of applying detailed models to a large number of sites, other factors, such as salt water intrusion and increased flood hazards were not examined. Estimating anticipated shoreline retreat and predicting the costs involved in holding back the sea, however, were deemed feasible goals. Physical Effects of Relative Sea Level Rise A rise in sea level will result in shoreline recession. The EPA estimated that a 1-meter (3.3 foot) rise in sea level would inundate 5,000 to 10,000 square miles (12,950 to 25,900 square 14

24 Originally published October 1991 by the Federal Emergency Management Agency, Federal Insurance Administration kilometers) of dry land if attempts to stabilize the shoreline are not made (Titus et al., 1989). Shoreline erosion is a worldwide problem; over 70 percent of the coastlines are undergoing significant erosion (Bird, 1985). Shoreline changes vary from the short-term erosion associated with individual storms to the longer-term effects of sea level rise. A significant rise in sea level establishes a setting in which increased erosion can occur. Land loss and barrier island submergence result from a combination of factors associated with relative sea level rise. Barrier islands are dynamic features which will respond to rising sea levels in various ways. Traditionally, barrier islands were thought to migrate landward due to the formation of inlets and overwash processes during storm events, allowing sand to be transported from the beach to the bay shore. However, it has been suggested that in the short term, many coastal barriers are actually eroding on both the beach and bay sides and essentially are being forced to drown in place (Leatherman, 1983). The NRC (1987) reports that shoreline erosion is probably responsible for about 1 percent of the total annual marsh losses. Land losses in marsh areas due to sea level rise are more commonly a result of ponding, the rapid enlargement of interior ponds in marshes which occurs if there is a large increase in sea level. Shoreline stabilization, e.g., bulkheads and levees, will affect the amount of marsh area lost to sea level rise by limiting the marsh's natural ability to trap sediments and build above the rising sea level. A change in the location and extent of coastal floodplain areas is another result of a rise in sea level. The change in floodplain area is dependent on slope, topography, use of protective coastal structures, and the magnitude of relative sea level rise. As the shoreline retreats in response to rising sea levels, additional areas of the floodplain will be submerged, and new areas will be periodically flooded. It is difficult to assess the extent of change in overall floodplain area due to a rise in sea level. However, the assumption can be made that an increase in relative sea level will result in new areas being subjected to the possibility of inundation by flood waters. Conversely, the resulting increase in shoreline erosion and submergence will cause a decrease in the area subject to flooding 3. The protective benefits offered by coastal structures will decline as sea level rises. Higher surge elevations and greater wave heights associated with an increase in sea level will result in an increase in destructive force and a decrease in the protection provided by the structure. Seawalls and bulkheads designed to withstand present estimates of wave action associated with a 100-year storm could be overtopped during storms of lesser magnitude, which could result in structural failure. Seawalls and bulkheads, which are often used to stabilize eroding shorelines and other areas 3 FEMA defines flooding to be "a general and temporary condition of partial or complete inundation of normally dry land" (44 CFR, Part 59). 15

25 Projected Impact of Relative Sea Level Rise on the National Flood Insurance Program vulnerable to wave attack, are usually not built to account for a significant short-term rise in sea level. The NRC (1987) suggests two ways that sea level rise could be incorporated in the design of coastal structures. Seawalls, bulkheads, and groins could be designed to accommodate the anticipated rise in sea level within the design life of the structure. Another method would be to upgrade the structure as sea level changes. Based on current estimates of sea level rise over the next century, structures with a design life of less than 50 years need not account for anticipated sea level rise (NRC, 1987). For a period of less than 50 years, modest increases in sea level based on current rates would amount to only a few inches and would therefore have little effect on most coastal structures. A secondary effect associated with sea level rise is an increase in coastal flooding due to the potential inundation of drainage systems beyond design capacity. Titus et al. (1987) examined the cost of constructing coastal drainage systems to accommodate a potential rise in sea level versus the retrofit cost of modifying existing structures if sea level rises. Their research indicates that retrofit costs depend on the type and design life of the existing structure, which varies from location to location, as well as on the overall change in sea level. 2.2 Purpose of Study The purpose of this study is to assess the implications of sea level rise (both physical and economic) on the NFIP. To accomplish this goal, analyses were performed to estimate changes over time in floodplain location and extent, and population density. In addition to these analyses, this study applied relevant results obtained from previous studies (NRC, 1987; EPA, 1989; IPCC, 1990) to help evaluate the overall impact of sea level rise on the NFIP. A nationwide assessment such as this study cannot incorporate detailed information on site-specific topography and different types of floodprone structures. However, general trends can be used to assess the potential impact on the NFIP. The task of predicting changes in relative sea level is a complex problem involving local, regional, and global factors. Undoubtedly, this complexity has led to the wide range of predictions concerning sea level change. Sea level rise during the next century will have a number of potential impacts on the NFIP. An evaluation of the effects includes consideration of flood risk assessment and flood insurance implications. The primary goals of the NFIP are the reduction of future flood losses and the transference of the costs of flood loss from the general taxpayer to those who choose to occupy floodplain areas. In support of NFIP goals, FEMA identifies flood risks and maps floodprone areas. The primary strategy adopted by the NFIP for reducing flood losses to new construction in identified floodprone areas is requiring, at a minimum, elevating and/or flood-proofing of new structures to the elevation of the flood that has a 1percent probability of being equaled or exceeded in 16